410 CONTRIBUTIONS FROM THE NATIONAL HERBARIUM. 



AUionia cordifolia (Kunze) Kuntze, Rev. Gen. PI. 533. 1891. 



Mentioned by Choisy and Hemsley as occurring within our limits, but the species 

 is a South American one and has not been collected in North America so far as our 

 herbarium material shows. 



Allionia linearifolia (S. Wats.) Kuntze, Rev. Gen. PI. 533. 1891. 



Oxybaphus linearifolius S. Wats. Proc. Amer. Acad. 17: 375. 1882. 



Allionia gratillima Standley, Contr. Nat. Herb. 12: 340. 1909. 



I have seen no specimens of this species from Mexico, but during the summer of 

 1910 Mr. J. C. Blumer collected excellent specimens in Arizona, only a few miles from 

 the boundary. As ho says in a letter received recently, the species will probably 

 be found farther south, on the Mexican side of the boundary. 



15. QUAMOCLIDION Choisy. 



Quamoclidion Choisy in DC. Prodr. 13 2 : 429. 1849. 



Type species, Mirabilis tn flora Benth. 



Quamoclidion as originally published consisted of two species, Q. nyctagineum and Q. 

 angulation. NyctagoanguiataBC. was quoted as a synonym of the second species, which 

 has never been identified. The specimens upon which it was based came from Mexico. 

 It has been suggested that the plant is AUioniella oxybaphoides, but the description 

 does not altogether agree with that genua nor has Allionella ever been found in Mexico 

 so far as can be determined. It is not improbable that it ia found in some of the moun- 

 tains of northern Chihuahua, but Choisy's specimens were collected by Mocino and 

 Sease" much farther south'. 



According to the description the characters of the plant are: Stem slender; leaves 

 petioled, cordate, acute, repand-angled; peduncles from axils of the upper leaves 

 scarcely surpassing them; involucres 5-lobed, 3-flowered, campanulate; flowers ses- 

 sile, reddish; perianth short-tubular, dilated and 5-toothed at the apex; stamens 

 long-exserted; stigma multifid; fruit obovate, not rugose. 



A tracing of Mocino and Sessd's drawing aeems not to agree with Choisy's descrip- 

 tion in all particulars. It represents a plant with 1-flowered involucres. The lobes 

 of the perianth are ovate and acute. In the dissection of the flower it is represented 

 as having 5 stamens, while i:i other flowers upon the branches the number of stamens 

 runs from 6 to 8. The drawing suggests an Allionia, but I have seen nothing that 

 resembles it even remotely. 



KEY TO THE SPECIES. 



Perianth 25 mm. long or less, with a very narrow limb 1. Q, triflorum. 



Perianth 40 to 50 mm. long, with a broad and conspicuous limb. 



Fruit dark brown to black; stems glabrous, at least below 2.Q. multiflorum. 



Fruit light brown, marked with 10 dark, vertical lines; stems 



usually pubescent throughout 3. Q.froebelii. 



1. Quamoclidion triflorum (Benth.) Standley, Contr. Nat. Aerb. 12: 358. 1909. 

 Mirabilis tri/lora Benth. PL llartweg. 23. 1839. 

 Quamoclidion nyctagineum Choisy in DC. Prodr. 13 2 : 429, 1849. 

 Type locality, Mexico. 

 Specimens examined: 



Baja California: From Cerro Colorado to Rodriguez, alt. 245 to 430 meters, 

 December 16, 1905, Nelson & Goldman 7324; from Miraflores to San Bernardo 

 Ranch in Sierra La Laguna, alt. 735 meters, January 20, 190G, Nelson & Gold- 

 man 7414; Triumfo, October 24, 1890, Brandegee 479; Pescadero, November, 

 1902, Brandegee; Todos Santos, 1890, Brandegee. 

 Mexico: No locality given, llartweg 197, type collection. 



